Different species with contrasting ecological demands react rather differently on the same landscape pattern (e.g., Louy et al. 2007). Our results indicated a higher structure for M. splendens than for O. curtisii, which agrees with the different dispersal of both species. However, these results should be taken as preliminary, due to the small sample size of some localities. For instance, Lérez was the locality where M. splendens (n=6) was the least genetically diverse, but the one with highest gene diversity and Shannon's index for O. curtisii (n=37). According to Aagaard et al. (1998), sample size might also explain the higher structuring of the two studied species compared to other Odonata (also surveyed using RAPDs) at similar geographical scale. Andrés et al. (2000) determined a low degree of structuring for Ischnura graellsii (6=0.061), whereas Andrés et al. (2002) found a moderate differentiation for Ceriagrion tenellum (6 = 0.142).
Different types of molecular markers may differ in the level of population differentiation they reveal (Volis et al. 2005). Therefore, it would be inappropriate to compare the extent of population structuring (as revealed by F analogs) we obtained for M. splendens and O. curtisii, with none or very little structuring found in Nehalennia irene (microsatellites; Wong et al. 2003), Calopteryx splendens (AFLPs; Svensson et al. 2004), Lestes viridis (allozymes, De Block et al. 2005), or Ischnura elegans (AFLPs; Abbott et al. 2008). In addition, comparisons with other studies should also include the geographical scale. The implications of a high population structuring (FST = 0.25-0.48) at fine scale (0-26 km), as shown by Lestes viridis (allozymes, Geenen et al. 2000) and Coenagrium mercuriale (microsatellites, Watts et al. 2004, 2006), might mislead the reconstruction of those processes leading to differentiation of populations as ours separated by 26.5-76.78 km.
To conclude, the present study determined that M. splendens is likely to be strongly structured in NW Iberia, whereas gene flow between different river basins may be the reason for the lower differentiation inferred for O. curtisii. Based on these results, we agree with the need of special management for M. splendens. Estimates of population size and assessment of putative connectivity networks between localities are needed for a better comprehension of the conservation status of O. curtisii.
Acknowledgements M. Pimentel and one anonymous referee improved the earlier versions of the manuscript with their comments. R. Sánchez-Guillén kindly helped in the laboratory. This work is part of MAA's PhD thesis. D. Romero designed Fig. 1. Funding was provided by grants from Spanish Ministry ofScience (B0S2001-3642) and Xunta de Galicia (PGIDT01MAM37101PR) to ACR. MV was financed by the Isidro Parga Pondal program and the University of A Coruna.
Was this article helpful?